Circular RNAs: novel regulators of dendritic protein synthesis during mammalian synapse development

神经元树突局部蛋白质合成的精确调控对于活性依赖的突触形成、重塑和可塑性是必不可少的，而这一过程中的缺陷与神经发育障碍有关，包括智力低下和自闭症。过去十年的研究发现，microRNAs(MiRNAs)是一类广泛的非编码RNA，RNA结合蛋白(RBP)是神经元局部蛋白质合成和树突形态发生的重要调节因子。然而，miRNA/RBP功能本身是如何由神经元活动来协调的，在很大程度上是未知的。在第一个资助期，我们重点研究了miRNA/RBP在调节树突形态中的作用。因此，我们确定了两个限制性商业惯例，Nova1和Ncoa3，它们是树突状miRNAs抑制活性所必需的。虽然Nova1是神经元miRISC的一部分，但Ncoa3通过诱导Ago2表达间接刺激miRISC功能(STörchel等人，EMBO J.2015)。我们进一步确定了一个竞争的内源RNA，Ube3a-1，作为一种额外的非编码调节机制，控制特定树突状miRNAs的可用性(Valluy等人，NAT。Neurosci 2015)。在第二个资助期，我们计划将重点放在环状RNA(CircRNAs)上，这是由特定的头尾剪接事件(反向剪接)产生的长非编码RNA的一个子类。在初步实验中，我们发现大鼠海马神经元中的CircRNAs含量丰富，高度富含在树突室，并受神经元活动的调节。RNAi介导的特定CircRNAs的敲除削弱了BDNF依赖的树突发生，表明它们参与了神经回路的发育。电子计算机分析表明，在特定的树突环状RNA中存在翻译起始位点和RBP结合基序，为进一步的机制研究提供了理论基础。总之，这导致我们假设CircRNAs是树突状蛋白合成和活性依赖的突触发育的重要调节因子。为了解决这一假说，我们将追求以下具体目标：(1)全面验证RNAseq确定的候选CircRNAs在体内外的树突定位；(2)利用RNAi技术询问多达24个候选CircRNAs在活性依赖突触发育和海马神经元可塑性中的功能；(3)通过长RNA-seq方法揭示树突状CircRNAs的内部序列和结构；(4)确定树突状CircRNA功能的机制，重点是miRNA/RBP海绵、树突递送工具和mRNA陷阱。通过解决这些目标，我们将获得前所未有的深入了解CircRNAs在哺乳动物突触发育和可塑性中的作用。这将为未来的研究提供一个框架，在更生理的背景下解决完整大脑中CircRNA的功能。

The precise regulation of local protein synthesis in neuronal dendrites is essential for activity-dependent synapse formation, remodeling and plasticity, and defects in this process are linked to neurodevelopmental disorders, including mental retardation and autism. Research during the last decade has identified microRNAs (miRNAs), an extensive class of non-coding RNAs, and RNA-binding proteins (RBP) as important regulators of local protein synthesis and dendrite morphogenesis in neurons. However, how miRNA/RBP function itself is orchestrated by neuronal activity is largely uknown. During the first funding period, we focused the role of a miRNA/RBP interplay in the regulation of dendrite morphology. Thereby, we identified two RBPs, Nova1 and Ncoa3, that are required for the repressive activity of dendritic miRNAs. Whereas Nova1 is part of neuronal miRISC, Ncoa3 indirectly stimulates miRISC function by inducing Ago2 expression (Störchel et al., EMBO J. 2015). We further identified a competing endogenous RNA, Ube3a-1, as an additional non-coding regulatory mechanism that controls the availability of specific dendritic miRNAs (Valluy et al., Nat. Neurosci 2015). In the second funding period, we plan to focus on circular RNAs (circRNAs), a subclass of long non-coding RNAs that are generated by specific head-to-tail splicing events (backsplicing). In preliminary experiments, we found that circRNAs in rat hippocampal neurons were abundant, highly enriched in the dendritic compartment and regulated by neuronal activity. RNAi-mediated knockdown of specific circRNAs impaired BDNF-dependent dendritogenesis, suggesting their involvement in neural circuit development. In silico analysis indicated the presence of translational start sites and RBP binding motifs in specific dendritic circRNAs, providing a rationale for further mechanistic studies. Together, this led us to hypothesize that circRNAs are important regulators of dendritic protein synthesis and activity-dependent synapse development. To address this hypothesis, we will pursue the following specific aims: (1) to comprehensively validate dendritic localization of candidate circRNAs identified by RNAseq in vitro and in vivo; (2) to interrogate the function of up to 24 candidate circRNAs in activity-dependent synapse development and plasticity of hippocampal neurons using RNAi; (3) to unravel the internal sequence and structure of dendritic circRNAs by long RNA-seq approaches; (4) to identify the mechanism underlying dendritic circRNA function, with a focus on miRNA/RBP sponges, dendrite delivery vehicles and mRNA traps. By addressing these aims, we will obtain unprecedented insight into the role of circRNAs in mammalian synapse development and plasticity. This will provide a framework for future studies that will address circRNA function in a more physiological context in the intact brain.